Levels and Trends in Absolute Poverty in the World: What we know and what we don't

This paper critically reviews the recent changes to the Global Poverty numbers generated by the World Bank in 2008. While they have little impact on observed poverty trends and while there are good reasons to believe that the previous numbers were on weak foundations, the new numbers on levels of poverty in the developing world create new uncertainties und questions. In particular, there are conceptual issues involved in using one ICP round to update all poverty numbers and more empirical issues related to the particular results of the 2005 ICP round. This paper reviews these issues and finds that we cannot be very certain about levels of absolute poverty in the world and that the current method for generating these absolute poverty numbers is problematic and should possibly be abandoned. At the same time, poverty trends are much less affected by these methodological issues. The paper also discusses potential alternatives to the current methods and highlights their strengths and weaknesses. Unfortunately, there is no readily available alternative to the current method, though with some difficulty, such an alternative could be developed.